Letters to MoneyWeek: tread carefully with statistics

Regarding your article "Insights that won a Nobel" (MoneyWeek 866), while it is time indeed that the myth of Homo economicus was laid to rest, it is prudent nonetheless to approach the results of behavioural psychology research with some degree of caution. Much recent experimentation in the field has been called into question due to the widespread and professionally sanctioned practice of "p-hacking" the unethical manipulation of data in search of statistical significance.

Stuart Vyse, author of Believing in Magic; the Psychology of Superstition and winner of the William James Book Award for the American Psychological Association, writes in the September issue of Skeptical Inquirer, "most studies need to report statistically significant results to have any chance of getting published so everyone is on a quest to achieve the vaunted p (for probability) < (less than) .05 that indicates the findings are unlikely to have happened by chance".

He continues: "Recent research aimed at trying to replicate previously published psychology studies has demonstrated shockingly that a large number of classic phenomena cannot be reproduced, and the popularity of p-hacking is thought to be one of the culprits."

As for Daniel Kahneman's Thinking, Fast and Slow, one encounters on page 57 of the book the following statement concerning psychological studies referred to earlier in the chapter, "You have no choice but to accept that the major conclusions of these studies are true." Such a statement is the antithesis of the scientific method and utterly alien to the principles of scientific inquiry. That a behavioural economist can take such a position in a published work is, to my mind, an indication of how far economics is from being anything like a science and economists anything like scientists.

It seems obvious, where investing is concerned, that a more deliberate and analytical approach will be likely to win out over gut instinct. That certainly has been our investment club's experience. We began the club in 2000 and by 2002 our unit value had dropped from its starting point of 100p to 45p. This prompted a change from a strategy of hunches, hot tips and gut instinct to one of solid research and analysis. Our unit value currently stands at 168p.

It is to Thaler's credit that he acknowledges that the overdue refutation of Homo economicus is nothing new, but rather a return of economics to its earliest roots.

MR

The problem of spurious results is a growing one we've written before about the number of smart-beta strategies that fall into this trap. We share your scepticism about much of this research and agree that the value of work by researchers like Thaler and Kahneman is that it is rooted in common sense.

Don't play down gambling dangers

Your article "Is gambling out of control in Britain?" (MoneyWeek 685) makes the false claim that only 0.5% of the population are problem gamblers. The most recent data from 2015 shows this figure to be 0.8%. However, the licensing objective of the 2005 Gambling Act is prevention of harm to the young and vulnerable, meaning that the 3.9% of the population who are at-risk gamblers should also be considered, making a total of 4.7% of the population.

In respect of bookies' fixed-odds betting terminals (FOBTs), at which over 3% of the population gamble, the statistics are 11.5% problem gamblers and 31.7% at-risk for a total of 43.2% vulnerable gamblers.

Sadly, headline data alone does not provide details on the depth of harm. Increases in debt, depression and divorce are in part attributable to increased gambling. FOBTs are the most dangerous form of gambling, with around 20% per year smashed by FOBT addicts.

While the reporter did provide some outlines on Campaign for Fairer Gambling positions, we were not contacted for comment. If we were, there might have been a more informative article and sub-title text of "A genuine problem or a moral panic?" might have been excluded.

Derek Webb Campaign for Fairer Gambling

MoneyWeek's usual policy is to publish letters anonymously in order to preserve readers' privacy, but in this case we are including the identity of the author in order to provide greater context.

In our view, this article gave a balanced overview of the issues surrounding gambling and did not suggest that concerns about FOBTs constituted a moral panic. There is clearly evidence to suggest that these terminals are damaging and help to facilitate gambling addiction. Consequently, it seems likely that they will be subject to greater regulation in future, regardless of what the bookies would wish.

The broader figure of "around 0.5%" of the population being problem gamblers is a general summary of the results of multiple studies conducted in the past few years, although of course some individual studies have put it at a slightly higher level. Regardless of the exact level, it does not suggest a widespread issue that justifies sweeping restrictions.

Hard Brexit means dearer food

In his piece "How to thrive during a no deal' divorce" (MoneyWeek 866), Matthew Lynn suggests that leaving the customs union would offer instant lower food prices because the UK would be able to lower tariffs on food products from third countries.

While long term this is a distinct possibility, in the immediate aftermath of the UK leaving the EU the opposite would be true. The UK government has signalled that it will initially keep the current EU tariffs on food imports, and they would therefore also apply to European imports in the event of a no-deal Brexit.

As up to 30% of all food consumed in the UK is imported tariff-free from the EU, these imports would immediately become much more expensive. The UK wouldn't be able to benefit from cheaper imports by unilaterally reducing tariffs for some time, and only after the tariff-rate quota issue had been resolved at the WTO.

In the long term there is an opportunity to move toward the type of system that Matthew Lynn proposes, but in the aftermath of a no-deal Brexit, food prices are more likely to rise than to go down.

DD

Writing to MoneyWeek

MoneyWeek welcomes letters and emails from readers, but unfortunately we are not able to publish or reply to all of them. We may edit letters prior to publication. All responses are for information only and should not be relied upon in making investment decisions.

Our staff are unable to respond to personal investment queries, as MoneyWeek is not authorised to provide individual investment advice. Please email us at editor@moneyweek.com, or write to us at Editor, MoneyWeek, 31-32 Alfred Place, London, WC1E 7DP.